Impact of Three-phase Pseudo-measurement Generation from Smart Meter Data on Distribution Grid State Estimation

This paper addresses the modelling of unknown power values, so-called pseudo-measurements, for the use in distribution grid state estimation. Historical load data from smart metering devices in private households and photovoltaic systems is analysed regarding phase distribution, reactive power and correlation between different generators. Based on this, methods for pseudomeasurement generation are developed and tested in combination with state estimation. Depending on the active power a corresponding value for the power factor has been found that enables the modelling of reactive power by using lookup tables. The high correlation in relative PV feed-in can be represented by assumption of equal standardized power. The developed methods are validated by simulation on the basis of an electric distribution grid. It is shown that phase distribution in pseudo-measurements is of minor importance, while the developed methods improve the quality of the distribution system state estimation result.